Campaign results provide you with all of the data points you need to make the right decisions, whether you are validating a concept or optimizing a live app. Here’s a quick rundown of all of the data points.
Overview
The overview tab shows each variant’s conversion rate, conversion rate distribution, and daily significance rate.
The probability of each variant winning again is calculated by utilizing 2 different statistical models: one is the Frequentist model and the other one is the Bayesian model. The Frequentist model calculates the frequency of occurrence of an event, observed in a number of repetitions of the experiment. The Bayesian model tells you how your current belief should be updated based on the new data that emerged.
If a clear winning variant has been found, it will be shown on the top. However, a winner is not always found, which can be due to many reasons, with the most common ones being:
- The differences in each variant’s performance are not clear enough, leading to our algorithm not having enough confidence in a winner
- The campaign did not get enough impressions for accurate data
- More than one winner was found
You can toggle variants by clicking to compare them to each other. Winning probabilities will change accordingly, so you can see what was the 2nd best performing, 3rd best, etc.
Performance over time
Performance over time shows the page views, installs and conversion rates of each variant. Toggle the variants by clicking on them. View them also as either cumulative or daily performance.
User segments
User segments show you the split between decisive and exploring users on your campaign and each variant. You can toggle to show only page views or installs, in addition to toggling by variant. These data points give valuable insight into the behavior of your audience.
Visual
This tab allows you to further see the differences between each variant. You can inspect the conversion rate per variant in the line chart. Additionally, if the look-alike page includes a video, you are able to see the watch rate and conversion rate per video. The conversion rate per video is derived by dividing the number of installs by the number of individuals who saw the specific video to a specific second.
Scroll heatmaps
This is the part where you get to see how the users behave on your store page, so where on the page does the user spend the most time. The higher the scroll engagement score, the better.
Clicks
A breakdown of the number of clicks is displayed to give data on what the user is doing on the look-alike page. Clicks are categorized as either neutral, positive or negative clicks:
- Neutral clicks don’t really affect the users behaviour on the page, i.e. clicking the icon
- Positive clicks are clicks that make the user stay on the page, i.e. the share button
- Negative clicks are clicks that drive the user away from the page, i.e. back button
Ad campaign & creative data
The multi-funnel introduces a new section at the bottom of the campaign results view. Now you are able to tag/edit your tracking links (both campaign and variation tracking links) with two additional parameters, campaign & ad. The impressions get tagged with these parameters, and allow you to view the results at the bottom of the campaign results page, by these parameters.
